Type 1 diabetes (T1D) is a complicated disease requiring constant adherence to glycemic targets and meticulous attention and vigilance of blood sugar. Although it is increasing in incidence the average life expectancy is 69 years, a 15 year increase from age 54 observed two decades ago. Concomitant with increased longevity is survival to an age at high risk of cognitive impairment and dementia both which greatly impact diabetes management and self-care. Elderly type 2 diabetes patients (T2D) have double the risk of cognitive impairment than normoglycemic individuals, yet it's unknown whether older T1D individuals are also at greater risk, and what the contributing and modulating factors are. T1D children, adolescents and middle- aged adults show deficits in certain domains of cognitive function (5), yet there's a lack of information regarding cognition in individuals aged 60 or older In contrast to elderly T2D patients, T1D have a much younger age of onset and longer duration of diabetes, increased hypoglycemia, more microvascular but less macrovascular complications, and by definition, are homogeneously treated with insulin. In the T2D population, hypoglycemic episodes, depression and microvascular comorbidities are established risk factors for cognitive impairment. Yet, it is unknown how these factors (2-3 times more common in T1D) influence cognitive aging in elderly T1D populations. We propose an innovative longitudinal study of cognitive aging in elderly T1D which will close the gap in knowledge regarding late-life cognitive function in this rapidly increasing group. We will exploit the exquisite data resources o the Kaiser Permanente of Northern California (KPNC) Diabetes Registry. The registry comprises over 147,000 elderly individuals with diabetes aged 60 and older (>7000 potential individuals with type 1 diabetes) from which to sample and prospectively investigate neuropsychological function and the profile of cognitive changes in this understudied and unique group. We will recruit 1200 patients aged 60-80 years (600 T1D and 600 without diabetes) and follow them over 4 years in a matched cohort study. At baseline and 36 months later a detailed neuropsychological battery will be given, and supplemented by annual phone administered collection of global cognition. Our goals are the following:1). Determine baseline and three year changes in global and domain specific cognition in elderly type 1 diabetes patients in comparison to a matched group of elderly individuals without diabetes, 2.) Characterize the role of glycemic control and vascular complications on cognitive change in the context of type 1 diabetes, and 3.) Evaluate the interaction of depression and type 1 diabetes on cognitive changes. This would be the first longitudinal study of cognitive aging in type 1 diabetes and has profound clinical implications for glycemic treatment targets and self-care. It will also provide a mechanistic framework for understanding glycemic and insulin related mechanisms in prevention of cognitive impairment and dementia